The Research Process in Nursing

By Anne Lacey

‘The perfect text for any health care professional who wishes to gain a sound understanding of research…This text succeeds where others fail in terms of the thoroughness of the research process and the accessible style in which the material is presented. In an age when nursing and health care research is going from strength to strength this book offers those in the world of academia and practice an excellent and essential 'bible' that is a must on any bookshelf’
Dr Aisha Holloway, Lecturer Adult Health, Division of Nursing, The University of Nottingham

‘a book that helps you each step of the way. A very understandable and enjoyable publication’
Accident and Emergency Nursing Journal

‘key reference resource that students of research can use at various levels of study. It is comprehensive, user friendly and very easy to read and make sense of’
Gillian E Lang, Amazon reviewer

The sixth edition of this book reflects significant developments in nursing research in recent years, ensuring the reader is provided with the very latest information on research processes and methods. It continues to explore how to undertake research as well as evaluating and using research findings in clinical practice, in a way that is suitable for both novice researchers and those with more experience.

Divided into six sections, the chapters are ordered in a logical fashion that also allows the reader to dip in and out. The first two sections of the book provide a comprehensive background to research in nursing. The third section presents a variety of qualitative and quantitative approaches,  both new and well-established. The final three sections then look at collecting and making sense of the resulting data and putting the research findings into clinical practice.

• Summarises key points at the start of each chapter to guide you through
• Includes contributions from a wide range of experts in the field
• Accessible but doesn’t shrink away from complex debates and technical issues

New to this edition:

• Accompanying website (www.wiley.com/go/gerrish)
• Ten completely new chapters including Narrative Research, Mixed Methods and Using Research in Clinical Practice
• ‘Research Example’ boxes from a wide variety of research types

• Language: English
• Publisher: Wiley
• Release date: Apr 2, 2013
• ISBN: 9781118682098

Book preview

SECTION 1

Setting the Scene

Nursing research does not exist in a vacuum, but is an applied discipline set in the context of a dynamic academic community and relating to a complex healthcare system. This section explores this context and introduces the reader to the nature of nursing research.

Chapter 1 presents the fundamental concepts of the discipline, reviews the current context of nursing research, and emphasises the essential connection between nursing research and the practice of the profession. Even those who do not see themselves as active researchers should be users of the knowledge generated by research, and so need to understand much of what follows in the sections of this book. Chapter 2 then takes the reader through the essential steps in the research process, each of which will be dealt with in much more depth in later sections, but with the aim of giving an overview of the entire undertaking that is research. Recent examples from the literature are used to illustrate the varied nature of nursing research.

Research in nursing, as in healthcare generally, is complicated by the fact that it is involved with vulnerable human beings, and ethical principles need to be observed from the outset of any research project. Chapter 3, therefore, tackles this moral obligation on the researcher, drawing out the practical implications for the researcher and setting the context for the more specific ethical regulations dealt with in Section 2 of the book.

The final two chapters in this section deal with the need for nursing research to be inclusive in scope. User involvement in research has been advocated from within and outside the profession for more than a decade now, and Chapter 4 argues for the full inclusion in the research process of those to whom the outcomes might apply. New to this edition of the book is Chapter 5 on research in a multi-ethnic society. Although there are many minority groups that deserve special consideration when designing nursing research, ethnicity perhaps merits particular consideration as a major factor impacting on healthcare in UK society.

CHAPTER 1

Research and Development in Nursing

Kate Gerrish and Anne Lacey

Key points

Research is concerned with generating new knowledge through a process of systematic scientific enquiry, the research process.

Research in nursing can provide new insights into nursing practice, develop and improve methods of caring, and test the effectiveness of care.

Whereas comparatively few nurses may undertake research, all nurses should develop research awareness and use research findings in their practice.

Evidence-based practice involves the integration of the best available research evidence with expert clinical opinion while taking account of the preferences of patients.

INTRODUCTION

Significant changes in healthcare have taken place in the 26 years since the first edition of this book was published and these changes are set to continue. Technological developments have led to improved health outcomes and at the same time have raised public expectations of healthcare services. Increased life expectancy and lower birth rates mean that the United Kingdom (UK) population is ageing. An older population is more likely to experience complex health needs, especially in regard to chronic disease, and this places additional demands on an already pressurised health service. At the same time, the escalating cost of healthcare is leading to a shift from expensive resource-intensive hospital care to more services being provided in the primary and community care sectors. In response to these changes, government health policy is increasingly focused on improving the clinical and cost-effectiveness of healthcare, while at the same time reducing the burden of ill health through active public health and health promotion strategies. For example, the recent review of the NHS in England undertaken by Lord Darzi has identified a number of priorities that need to be progressed in order to provide high quality care for patients and the wider public (DoH 2008). The review stresses the importance of improving health outcomes by preventing illness, as well as enhancing the quality of care provided to people with particular needs, for example patients with common long-term conditions such as diabetes, or those in need of palliative and end-of-life care.

To achieve the outcomes for enhancing quality set out in the review, there is a need to change the way healthcare professionals work and the way health services fit together, and to ensure that patients have access to the best available treatments. However, achieving quality in healthcare is a moving target. What was considered high quality care in 1948 when the NHS was first founded is no longer considered to be the case 60 years on. Knowledge about effective healthcare interventions has increased enormously, and this is certainly the case with nursing interventions. In the past, custom and practice, often based on the ward sister’s or doctor’s likes and dislikes, dictated what nurses did to patients, but nursing research has provided a new evidence base to inform the care that nurses provide. One clear example is in the field of pressure area care. It is not that long ago that nurses applied various techniques in an attempt to reduce the risk of a patient developing a pressure sore, these included egg white and oxygen, methylated spirits and vigorously rubbing the area at risk. Yet research by Doreen Norton more than 30 years ago clearly identified that moving patients regularly, keeping their skin clean and dry, and using the right equipment was the most effective way to reduce the risk (Norton et al. 1975).

It is essential that nurses respond proactively to the developments outlined above in order to provide high quality care in response to the needs of the individuals and communities with whom they work. To do this, they need up-to-date knowledge to inform their practice. Such knowledge is generated through research. This chapter introduces the concept of nursing research and considers how research contributes to the development of nursing knowledge. In recognising that nursing is a practice-based profession the relevance of research to nursing policy and practice is examined within the context of evidence-based practice and the responsibilities of nurses is explored in respect of research awareness, research utilisation and research activity.

NURSING RESEARCH AND DEVELOPMENT

The definition of research provided by Hockey (1984) in the first edition of this book is still pertinent today:

‘Research is an attempt to increase the sum of what is known, usually referred to as a body of knowledge by the discovery of new facts or relationships through a process of systematic scientific enquiry, the research process’ (Hockey 1984: 4)

Other definitions of research emphasise the importance of the knowledge generated through research being applicable beyond the research setting in which it was undertaken, i.e. that it is generalisable to other similar populations or settings. The Department of Health, for example, defines research as:

‘the attempt to derive generalisable new knowledge by addressing clearly defined questions with systematic and rigorous methods’ (DoH 2005: 3, section 1.10)

Research is designed to investigate explicit questions. In the case of nursing research these questions relate to those aspects of professional activity that are predominantly and appropriately the concern and responsibility of nurses (Hockey 1996). The International Council of Nursing’s (ICN) definition of nursing research captures the broad areas of interest that are relevant to nurse researchers.

‘Nursing research is a systematic enquiry that seeks to add new nursing knowledge to benefit patients, families and communities. It encompasses all aspects of health that are of interest to nursing, including promotion of health, prevention of illness, care of people of all ages during illness and recovery or towards a peaceful and dignified death’ (ICN 2009)

The ICN has identified nursing research priorities in two broad areas, namely health and illness, and the delivery of care services. These priority areas are outlined in Box 1.1. In addition, research in the field of nursing education is important, for unless nurses are prepared appropriately for their role, they will not be able to respond to the needs of patients, families and communities. Priorities for research in nursing education are broad ranging as illustrated in Box 1.2. Most nursing research investigates contemporary issues; however, some studies may take an historical perspective in order to examine the development of nursing by studying documentary sources and other artefacts (see Chapter 26).

The questions that nursing research may address vary in terms of their focus. More than 20 years ago, Crow (1982) identified four approaches that research could take; these remain pertinent today:

Box 1.1 Priorities for nursing research identified by the International Council of Nurses

Health and illness

Nursing research priorities in health and illness focus on:

health promotion

prevention of illness

control of symptoms

living with chronic conditions and enhancing quality of life

caring for clients experiencing changes in their health and illness

assessing and monitoring client problems

providing and testing nursing care interventions

measuring the outcomes of care.

Delivery of care services

Nursing research priorities in delivery of care services focus on:

quality and cost-effectiveness of care

impact of nursing interventions on client outcomes

evidence-based nursing practice

community and primary healthcare

nursing workforce to include quality of nurses’ work life, retention, satisfaction with work

impact of healthcare reform on health policy, programme planning and evaluation

impact on equity and access to nursing care and its effects on nursing

financing of healthcare.

Source: ICN (2009)

Box 1.2 Priorities for research in nursing education

Curriculum design and evaluation, including community-driven models for curriculum development

New pedagogies

Innovation in teaching and learning

Use of instructional technology, including new approaches to simulated learning

Student/teacher learning partnerships

Clinical teaching models

Assessment of student learning in classroom and practice settings

New models for teacher preparation and faculty development

Quality improvement processes

Educational systems and infrastructures

Adapted from National League for Nursing (2008)

research that will provide new insights into nursing practice

research that will deepen an understanding of the concepts central to nursing care

research that is concerned with the development of new and improved methods of caring

research that is designed to test the effectiveness of care.

Nursing research does not necessarily need to be undertaken by nurses. Indeed, some seminal studies into nursing practice and nurse education have been undertaken by sociologists. For example, in the 1970s, Robert Dingwall, a sociologist, undertook an influential study of health visitor training (Dingwall 1977). Likewise, nurses who engage in research may not confine their area of enquiry to nursing research. The growing emphasis on multidisciplinary, multi-agency working means that nurse researchers may choose to examine questions that extend beyond the scope of nursing into other areas of health and social care. Nurse researchers may find themselves working in multidisciplinary teams including statisticians, health economists, sociologists and other health professionals, working on research areas such as rehabilitation, which encompass a wide range of disciplines. Nurse researchers work appropriately in university departments such as social science, health services research and complementary medicine, as well as in departments of nursing and midwifery.

Whereas the generation of new knowledge is valuable in its own right, the application and utilisation of knowledge gained through research is essential to a practice-based profession such as nursing. This latter activity is known as ‘development’. Thus research and development, ‘R&D’, go hand in hand.

Research and development can be divided into three types of activity.

Basic research is original, experimental or theoretical work, primarily for the purpose of obtaining new knowledge rather than focusing on the specific use of the findings. For example, biomedical laboratory-based research falls within this category.

Applied research is also original investigation with a view to obtaining new knowledge, but it is undertaken primarily for practical purposes. Much nursing research falls within this category and is undertaken with the intention of generating knowledge that can be used to inform nursing practice and can involve both clinical and non-clinical methods.

Development activity involves the systematic use of knowledge obtained through research and/or practical experience for the purpose of producing new or improved products, processes, systems or services.

Development activity that focuses on the utilisation of knowledge generated through research can take different forms. The most common activities include clinical audit, practice development and service evaluation (see Box 1.3). Like research, these activities often employ systematic methods to address questions arising from practice. Research, however, is undertaken with the explicit purpose of generating new knowledge, which has applicability beyond the immediate setting. By contrast, clinical audit, practice development and service evaluation are primarily concerned with generating information that can inform local decision making (NPSA 2008). Yet, the boundaries between some forms of research, for example action research (see Chapter 22) and practice development, and evaluation research (see Chapter 21) and service evaluation, are often blurred (Gerrish & Mawson 2005). It is, however, important to be able to differentiate between these activities as they require very different approval processes before the work can begin (see Chapter 10).

DEVELOPING NURSING KNOWLEDGE

Nursing research is concerned with developing new knowledge about the discipline and practice of nursing. Nursing knowledge, like any other knowledge, is never absolute. As the external world changes, nursing develops and adapts in response. In parallel, nursing knowledge develops and changes. This year’s ‘best available evidence’ has the potential of being superseded by new insights and discoveries. Therefore nursing knowledge is temporal, and will always be partial and hence imperfect. This does not mean, however, that nurses should not continually strive to develop new knowledge to inform nursing and healthcare policy and practice.

Box 1.3 Definitions of research, clinical audit and practice development

Research involves the attempt to extend the available knowledge by means of a systematically and scientifically defensible process of inquiry (Clamp et al. 2004).

Clinical audit is a professional-led initiative that seeks to improve the quality and outcome of patient care through practitioners examining their practices and results and modifying practice where indicated (NHSE 1996: 16). Clinical audit measures care against pre-determined standards.

Practice development encompasses a broad range of innovations that are initiated to improve practice and the services in which that practice takes place. It involves a continuous process of improvement towards increased effectiveness in patient-centred care. This is brought about by helping healthcare teams to develop their knowledge and skills, and transform the culture and context of care (Garbett & McCormack 2002).

Service evaluation seeks to assess how well a service is achieving its intended aims. It is undertaken to benefit those who use a particular service and is designed and conducted solely to define or judge current service (NPSA 2008).

Whereas the focus of this book is on the generation of knowledge through research, it is important to recognise that nursing knowledge may take different forms. In addition to empirical knowledge derived through research, nurses use other forms of knowledge, such as practical knowledge derived from experience, and aesthetic or intuitive knowledge derived from nursing practice (Thompson 2000). Nurses use these different forms of knowledge to varying degrees to inform their practice (Gerrish et al. 2008). It is beyond the scope of this book to examine in detail the various forms of nursing knowledge; however, Chapter 38 introduces the reader to some of these within the context of promoting evidence-based practice.

The definitions of research given earlier in this chapter emphasise the role of systematic scientific enquiry – the research process – in generating new knowledge. The research process comprises a series of logical steps that have to be undertaken to develop knowledge. Different disciplines may interpret the research process in different ways, depending on the specific paradigms (ways of interpreting the world) and theories that underpin the discipline. A biological scientist’s approach to generating new knowledge will be different from that of a sociologist. However, the basic principles of the systematic research process will be followed by all disciplines. Nursing, as a discipline in its own right, is relatively young in comparison to more established professional groups such as medicine, and is in the process of generating theories that are unique to describing, explaining or predicting the outcomes of nursing actions. Nursing theories are generated through the process of undertaking research and may also be tested and refined through further research. However, nursing also draws on a unique mix of several disciplines, such as physiology, psychology and sociology, and any of these disciplines may be appropriate for research in nursing. For example, the management of pain can be studied from a psychological or physiological perspective; whichever approach is chosen will be influenced by the theories relevant to the particular discipline.

The research process in nursing is no different from that of other disciplines and the same rules of scientific method apply. Chapter 2 sets out a systematic approach to research – the scientific method in action – and subsequent chapters consider the various components of the research process in detail. At this stage, it is worth noting that in some texts, the ‘scientific method’ is taken to reflect a particular view of the world which values the notion that we can be totally objective in our research endeavours. Here, the term is not restricted in this way and we use the term ‘scientific method’ to mean a rigorous approach to a systematic form of enquiry. Chapter 11 introduces the reader to the different ways in which the scientific method can be interpreted, depending on the assumptions that the researcher holds about the nature of the social world and reality. These can be broadly classified as quantitative and qualitative approaches to research. Quantitative research is designed to test a hypothesis and generally involves evaluating or comparing interventions, particularly new ones, whereas qualitative research usually involves seeking to understand how interventions and relationships are experienced by patients and nurses (NPSA 2008).

RESEARCH AWARENESS, UTILISATION AND ACTIVITY

Research-based practice is arguably the hallmark of professional nursing and is essential for high quality clinical and cost-effective nursing care (ICN 2009). It is now more than 35 years since the Report of the Committee on Nursing (Committee on Nursing/Briggs report 1972) stressed the need for nursing to become research based to the extent that research should become part of the mental equipment of every practising nurse. Although considerable progress has been made in the intervening period, this objective still remains a challenge. For nursing to establish its research base, nurses need to develop an awareness of research in relation to practice, they need to be able to utilise research findings and some nurses need to undertake research activity.

Research awareness implies recognition of the importance of research to the profession and to patient care. It requires nurses to develop a critical and questioning approach to their work and in so doing identify problems or questions that can be answered through research. Nurses who are research aware will be able find out about the latest research in their area and have the ability to evaluate its relevance to practice. They will also be open to changing their practice when new knowledge becomes available. Research awareness implies a willingness to share the task of keeping abreast of new developments by sharing information with colleagues. It also entails supporting and co-operating with researchers in an informed way. Nurses need to understand the implications for patients that arise from research being undertaken in the clinical area in which they work. For example, nurses may need to provide care according to an agreed research protocol, and deviating from the protocol may jeopardise the research. However, they must ensure that the wellbeing of patients remains paramount and report promptly any concerns they may have about the research to more senior clinicians/managers as well as researchers. Arguably, all nurses should develop research awareness as part of pre-registration nurse education programmes and continue to develop their knowledge and skills following registration.

Research utilisation is concerned with incorporating research findings into practice so that care is based on research evidence. Not all research, even that which is published in reputable journals, is necessarily of high quality. Before findings can be applied a research study needs to be evaluated critically to judge the quality of the research. All nurses should be able to appraise a research report, although specialist advice may be needed to help judge the appropriateness of complex research designs or unusual statistical tests. Chapter 7 provides guidance on how to appraise research reports.

Research studies do not always provide conclusive findings that can be used to guide practice. Different studies examining the same phenomenon may produce contradictory results. Wherever possible a systematic review of a number of studies examining a particular phenomenon should be undertaken to provide more robust guidance for practice than the findings of a single study would allow. Chapter 24 outlines the procedures for undertaking a systematic review. It is a time-consuming process and requires a good understanding of research designs and methods together with knowledge of techniques for analysis, including statistical tests. Whereas some nurses may develop the skills to undertake a systematic review as part of a postgraduate course, many systematic reviews are undertaken by people who are experts in the technique. For example, the Centre for Reviews and Dissemination at the University of York has been set up specifically for the purpose of undertaking systematic reviews on a range of health-related topics.

The findings from a systematic review then need to be incorporated into clinical guidelines or care protocols that can be applied to practice. Whereas some guidelines may be developed at a national level, nurses may need to adapt national guidelines for application at a local level or develop their own guidelines where no national ones are available (see Chapters 38 and 39 for more information).

All nurses should be research aware and use research findings in their practice; however, not all nurses need to undertake research. To carry out rigorous research, nurses need to be equipped with appropriate knowledge and skills. Pre-registration and undergraduate post-registration nursing programmes tend to focus on developing research awareness and research utilisation. It is generally not until nurses embark on a master’s programme or a specialist research course that they will learn how to undertake a small-scale research study under the supervision of a more experienced researcher. This represents the first step in acquiring the skills to become a competent researcher. Comparatively few nurses progress to develop a career in nursing research in which they undertake large-scale studies funded by external agencies. The ability to lead a large-scale study generally requires study at doctoral level, followed by an ‘apprenticeship’ working within a research team with supervision and support from experienced researchers. A recently published report on clinical academic careers for nurses provides a framework to enable nurses to develop their competence as researchers while still maintaining and developing their clinical role. The new clinical academic training pathway creates opportunities for nurses to progress from master’s programmes in clinical research, through doctoral and post-doctoral clinical research opportunities with the ultimate aim of holding a senior clinical academic appointment between a university and an NHS trust (UKCRC 2007).

Although relatively few nurses progress to lead large research studies, many more nurses participate in research led by nurse researchers, doctors and other health professionals. Nurses working in clinical practice may be asked to undertake data collection for other researchers, and their clinical nursing experience can be valuable to the research enterprise. Even if they are not leading a study, nurses who assist other researchers should have a sound understanding of the research process in order to collect valid and reliable data and to adhere to the research governance and ethical requirements outlined in Chapter 10.

RESEARCH AND NURSING PRACTICE

Current policy initiatives seek to promote a culture of evidence-based practice. There are generally considered to be three components to evidence-based practice, namely the best available evidence derived from research, clinical expertise and patient preferences (Sackett et al. 1996). In recognising that knowledge derived from research is never absolute, nurses should draw on their own expertise and that of other more experienced nurses when deciding on an appropriate intervention. Equally, clinical expertise should not be seen as a substitute for research evidence, but rather as contributing to the decision about the most appropriate intervention for a particular patient. The third component of evidence-based practice involves taking account of patient perspectives. Nurses have a responsibility to share their knowledge of the best available evidence with patients to help them make informed choices about the care they receive. This is particularly important where there are alternative courses of action that can be selected. These issues are examined in more detail in Chapters 38 and 39.

Nursing’s progress towards becoming evidence based needs to be viewed within the context of wider influences on healthcare. The UK (England, Northern Ireland, Scotland and Wales) governments are each seeking to modernise the NHS through major policy reforms. Central among these initiatives has been the introduction of the concept of clinical governance, a process whereby healthcare organisations are accountable for continually improving the quality of their services and safeguarding high standards of care by creating an environment that promotes excellence (Currie et al. 2003). It is, however, difficult to achieve the aspiration of ‘excellence’ in healthcare because of financial constraints and pressure on resources (Maynard 1999). Nevertheless, the objective of seeking to develop the quality of healthcare, together with recognition of the importance of healthcare organisations and the individuals who work in them being accountable for the quality of services, is laudable. Research is essential for making progress towards achieving this objective. As outlined earlier in this chapter, the knowledge generated through nursing research should be used to develop evidence-based practice, improve the quality of care and maximise health outcomes (ICN 2009).

In order to enhance the quality of nursing care it is important to ensure that care is clinically effective. Often referred to as ‘doing the right thing right’, clinical effectiveness involves providing the most appropriate intervention in the correct manner at the most expedient time, in order to achieve the best outcomes for the patient. Nurses need to draw on knowledge generated through research to decide which intervention is most appropriate and how and when to deliver it. Research may also highlight reasons for non-compliance. For example, a particular dressing may have been shown through research to be effective in promoting wound healing, but if it is unacceptable to the patient problems with compliance may arise.

As mentioned earlier, the findings from a single study may not provide sufficient evidence to direct practice, and wherever possible nurses should rely on knowledge generated through systematic reviews of research evidence drawn from several research studies. There are a number of national initiatives to assist nurses and other health professionals, to provide clinically effective care. These include the development of clinical guidelines based on the best research evidence by, for example, the National Institute for Health and Clinical Excellence (NICE) and the Scottish Intercollegiate Guideline Network (SIGN). In addition, the recently launched NHS Evidence portal provides healthcare professionals with access to a comprehensive evidence base to inform clinical practice. It is intended to provide a ‘one-stop shop’ for a range of information types, including primary research literature, practical implementation tools, guidelines and policy documents (see the list of websites at the end of the chapter).

Increasing demands on the finite resources within the NHS have resulted in the need to ensure that healthcare interventions are not only clinically effective but also cost-effective. There is little point pursuing a costly intervention if a cheaper one is seen to be equally as effective. The field of health economics is concerned with examining the financial and wider resource implications of providing a specific intervention or service. Economic evaluations can be undertaken to evaluate different treatments or alternative ways of providing services from an economic perspective and providing information that can be used to inform judgements about the clinical and cost-effectiveness of a particular intervention or service (Chambers & Boath 2001). NICE and SIGN guidelines take account of both clinical and cost-effectiveness when making recommendations for best practice.

CONCLUSIONS

Research is necessary to develop the knowledge base to inform nursing policy and practice. In an era of evidence-based practice, nurses are constantly challenged to identify new and better ways of delivering care that is grounded in knowledge derived from research (ICN 2009). They have a professional obligation to their patients and to wider society to provide care that is based on the best available evidence. Whereas relatively few nurses will develop a career in nursing research, all nurses should become research aware. This means developing a critical and questioning approach in order to identify areas where practice could be improved on the basis of research findings or areas where research evidence is lacking and new knowledge needs to be generated through research. Nurses also need to utilise research findings in their day-to-day practice. However, in order to provide evidence-based care nurses need to be able to evaluate the quality of published research reports. This requires a sound understanding of the research process, together with knowledge of different research designs and the methods that can be used to collect and analyse data. The following chapters of this book examine the research process, designs and methods in detail in order to equip nurses with the knowledge base to critically appraise research reports and to engage in the process of undertaking research under the supervision of a more experienced researcher.

References

Chambers R, Boath E (2001) Clinical Effectiveness and Clinical Governance Made Easy, 2nd edition. Abingdon, Radcliffe Medical Press.

Clamp C, Gough S, Land L (2004) Resources for Nursing Research: an annotated bibliography, 4th edition. London, Sage.

Committee on Nursing (1972) Report of the Committee on Nursing. London, HMSO.

Crow R (1982) How nursing and the community can benefit from nursing research. International Journal of Nursing Studies 19(1): 37–45.

Currie L, Morrell CJ, Scrivener R (2003) Clinical Governance: an RCN resource guide. London, Royal College of Nursing.

Department of Health (2008) High Quality Care for All: NHS next stage review final report. Cm. 7432. London, The Stationery Office.

Department of Health (2005) Research Governance Framework for Health and Social Care, 2nd edition. London, Department of Health.

Dingwall R (1977) The Social Organisation of Health Visitor Training. London, Croom Helm.

Garbett R, McCormack B (2002) A concept analysis of practice development. Nursing Times Research 7(2): 87–100.

Gerrish K, Ashworth P, Lacey A, Bailey J (2008) Developing evidence-based practice: experiences of senior and junior clinical nurses. Journal of Advanced Nursing 62: 62–73.

Gerrish K, Mawson S (2005) Research, audit, practice development and service evaluation: implications for research and clinical governance. Practice Development in Health Care 4(1): 33–39.

Hockey L (1996) The nature and purpose of research. In: Cormack DFS (ed) The Research Process in Nursing, 3rd edition. London, Blackwell Science, pp 3–13.

Hockey L (1984) The nature and purpose of research. In: Cormack DFS (ed) The Research Process in Nursing, 1st edition. London, Blackwell Science, pp 1–10.

International Council of Nurses (2009) Nursing research: a tool kit. Geneva, International Council of Nurses. www.icn.ch/matters_research.htm (accessed July 2009).

Maynard A (1999) Clinical governance: the unavoidable economic challenges. NT Research 4: 188–191.

National Health Service Executive (1996) Promoting clinical effectiveness: a framework for action. London, NHSE.

National League for Nursing (2008) Research Priorities for Nursing Education. New York, National League for Nursing. www.nln.org/research/priorities.htm (accessed July 2009).

National Patient Safety Agency (2008) Defining Research. London, National Patient Safety Agency.

Norton D, McLaren R, Exton-Smith A (1975) An Investigation of Geriatric Nursing in Hospital. London, Churchill Livingstone.

Sackett DL, Rosenburg WM, Muir Gray JA, Haynes RB, Richardson WS (1996) Evidence-based medicine: what it is and what it isn’t. British Medical Journal 312: 71–72.

Thompson DR (2000) An exploration of knowledge development in nursing – a personal perspective. NT Research 5(5): 391–394.

UKCRC Subcommittee for Nurses in Clinical Research (Workforce) (2007) Developing the Best Research Professionals. Qualified Graduate Nurses: recommendations for preparing and supporting clinical academic nurses of the future. London, UKCRC.

Websites

www.evidence.nhs.uk – NHS Evidence website provides access to a comprehensive evidence base to inform clinical practice. It provides a ‘one-stop shop’ for a range of information types, including primary research literature, practical implementation tools, guidelines and policy documents.

www.nice.org.uk – National Institute for Health and Clinical Excellence (NICE) publishes recommendations on treatments and care using the best available evidence of clinical and cost-effectiveness.

www.rcn.org.uk/development/researchanddevelopment – RCN Research and Development Co-ordinating Centre website provides links to a range of resources to support nursing research and evidence-based practice.

www.sign.ac.uk – Scottish Intercollegiate Guidelines Network (SIGN) publishes national clinical guidelines containing recommendations for effective practice based on current evidence.

www.york.ac.uk/inst/crd – Centre for Reviews and Dissemination (CRD) undertakes and publishes reviews of research about the effects of interventions used in health and social care.

CHAPTER 2

The Research Process

Anne Lacey

Key points

The research process is a series of steps that need to be undertaken to carry out any piece of research.

The precise stages of the research process, and the order in which they are undertaken, will vary depending on the nature of the research, but will always follow a systematic pattern from initial ideas through to dissemination and implementation.

Rigour in research is essential if the work is to be trustworthy and free from bias.

INTRODUCTION

The process of undertaking research is essentially the same whether the subject matter of the research is pure science, medicine, history or nursing. The following rather expansive definition from Graziano and Raulin (2004) sums up the breadth of scope of the research process:

‘Research is a systematic search for information, a process of inquiry. It can be carried out in libraries, laboratories, schoolrooms, hospitals, factories, in the pages of the Bible, on street corners, or in the wild watching a herd of elephants’ (Graziano & Raulin 2004: 31)

In all cases the researcher must ascertain the extent of existing knowledge, define their own area of enquiry, collect data and analyse it, and draw conclusions. For the pure scientist, however, the research might take place in the context of a laboratory, where experimentation is relatively straightforward as the researcher is in control of the environment and can eliminate potential confounding factors that might invalidate the research. Unless using animals or human tissue, there are few ethical considerations to take into account.

For the student of nursing research, or any research in a social context, the process is complicated by practical and ethical constraints of working in the ‘real world’ (Robson 2002). There is no single, universally accepted way of carrying out research in the social world, but a plethora of different designs and methodologies ranging from phenomenology to randomised controlled trials, from epidemiology to action research. The range of approaches derives from different paradigms, or ways of seeing the world. However, all are valid ways of conducting research, provided the methodology used is appropriate for the research question and is applied in a rigorous, systematic fashion.

In this chapter the research process that is common to all nursing research will be explored, and subsequent chapters in Section 2 will look at each of the stages of research in more detail. Different methodologies or research designs are discussed in turn, and in detail, in Section 3.

Although the research process will be presented as a linear, sequential process, the stages are often revisited several times during the process. In qualitative research, in particular, it is likely that the ‘stages’ of the research process will be modified to take account of the emergent nature of the enterprise. Qualitative researchers sometimes find it difficult or even inappropriate to formulate a precise research question until they have begun to collect, and possibly even analyse, data.

However, it is helpful in the first instance to think through the entire research process in a systematic way. Many authors (Hek et al. 2006, Parahoo 2006, Moule & Goodman 2009) have described the research process, and each comes up with a different number of stages, but essentially they contain the same elements. Table 2.1 illustrates the process as it will be described in this chapter, and indicates the principal chapters in the book that deal with each stage. This chapter gives a brief overview of the various stages to enable readers to see the whole before looking at each stage in more detail.

DEVELOPING THE RESEARCH QUESTION

Most research questions begin with a ‘hunch’ or initial idea that is not precisely defined. The idea might arise from clinical practice, from professional discussion among colleagues, from an issue in the media, or from reading an article or book. Alternatively the question may be derived from a ‘call for proposals’ from a funding body that asks researchers to develop a proposal on a specific topic. Box 2.1 provides an example of such a call, in this case from the National Institute for Health Research Service Delivery and Organisation (SDO) Programme. The call is specifically about the research areas to be investigated, indicates the methods to be used, the funding available and timescale required. Full details are available from the SDO website, together with a standard application form and a deadline by which proposals have to be submitted.

But most nurse researchers begin with an initial idea that is not yet well defined. Let us consider how research questions might be developed, using some real examples from the nursing literature to illustrate our discussion (see Research Examples 2.1, 2.2 and 2.3).

Question 1

Perhaps a research team has a ‘hunch’ that the use of pelvic floor exercises might help women in the second stage of labour. This hunch is probably based on knowledge of the anatomy of the pelvic muscles and the process of delivery. It might also be supported by personal or professional experience of midwives. There are several ways in which the question could be developed. The following are examples of research questions derived from this area of interest.

Table 2.1 The research process

Box 2.1 SDO call Patient and carer-centred services/technology adoption

Reproduced by permission of the NIHR Service Delivery and Organisation programme Call EV2001. © Queen’s Printer and Controller of HMSO 2009.

RESEARCH EXAMPLE 2.1 A Quantitative Experimental Study

Salvesen KA, Morkved S (2004) Randomised controlled trial of pelvic floor muscle training during pregnancy. British Medical Journal 329: 378–380.

This study used a quantitative experimental approach to assess the effectiveness of using a structured training programme of pelvic floor exercises in reducing time spent in the second stage of labour during childbirth. Researchers in Norway recruited 301 first-time mothers during pregnancy, and randomly allocated them to either a training group (who were given an exercise programme delivered by a physiotherapist) or a control group (who had normal care). Time spent in the second stage of labour was measured for the two groups. Results showed that women in the training group had a lower rate of prolonged second stage labour than women in the control group (25% compared to 33%).

RESEARCH EXAMPLE 2.2 A Quantitative Questionnaire Survey

Chevalier I, Benoit G, Gauthier M, Phan V, Bonnin A, Lebel M (2008) Antibiotic prophylaxis for childhood urinary tract infection: a national survey. Journal of Paediatrics and Child Health 44: 572–578.

A national survey of Canadian paediatricians was conducted to assess their practice in prescribing prophylactic antibiotics for children with urinary tract infections, with and without vesicoureteral reflux. A self-completion questionnaire was mailed to a sample of 1136 paediatricians and 42 paediatric nephrologists. A response rate of 58.1% was obtained. Although a majority of respondents prescribed prophylaxis for children with reflux, only 15% felt that this practice was evidence based. A quarter of respondents also prescribed prophylaxis for children under one year with a first febrile urinary tract infection without evidence of reflux. Again, only 19% felt that this practice was evidence based. The overall conclusion was that practice in this area varies widely in Canada because of a lack of solid evidence about prophylaxis.

RESEARCH EXAMPLE 2.3 A Qualitative Study

Hasson F, Kernohan W, Waldron M, Whittaker E, McLaughlin D (2008) The palliative care link nurse role in nursing homes: barriers and facilitators. Journal of Advanced Nursing 64: 233–242.

This descriptive qualitative study explored the views and experiences of link nurses for palliative care working in nursing homes in Northern Ireland. A purposive sample of 14 link nurses from 10 nursing homes was selected and interviewed using focus groups. Data from the focus groups were recorded, transcribed and analysed. Link nurses identified a number of barriers to their role as educators and facilitators of palliative care, including lack of management support, a transient workforce and lack of adequate preparation for the role. Facilitators included external support, peer support and access to a resource file. The researchers concluded that the link nurse role had considerable potential to improve care in this area, but managers needed to be aware of the sustained support needed for the role, and more work needs to be done to find ways of developing the role further.

Obviously, each of these research questions will give us very different kinds of information and will require different research methods to be employed. They would also need to be refined further – the precise pelvic floor exercises to be taught needs to be clarified, for example, and the stage of pregnancy at which they are taught needs to be defined. Q1(b) suggests the need to measure understanding and willingness to learn – neither of these concepts is straightforward and tools to measure them would need to be developed. Perhaps a qualitative study needs to be undertaken to explore the concepts first. Research Example 2.1 (Salveson & Morkved 2004) describes an experimental study related to Q1(c). In this case the outcome measure was defined as time taken in second stage of labour, and only first-time mothers were recruited to the study.

Question 2

Alternatively, a research team might be interested in the evidence base used by doctors in their prescribing practice. Overuse of antibiotics in children, for instance, is known to cause problems with the development of drug resistance, and it is important that clinical practice is based on sound clinical evidence. Again, a number of research questions could be asked.

Again, these three questions lead to very different types of study, and again, each question needs further clarification and refinement. What is meant by ‘urinary disease’? How do we decide that research evidence is reliable? What age children are concerned? Which children are ‘at risk’? Research Example 2.2 (Chevalier et al. 2008) is an example of a survey to answer Q2(c), but it was undertaken with a specific group of paediatric doctors in Canada. Is it appropriate to apply the answers gained from this study to doctors in Europe or China?

Question 3

In our last example, research questions might be generated concerning the best way to deliver palliative care in nursing homes. This setting is known to be a common one in which palliative care is delivered, but formal training and facilities are not always available. Three questions could be constructed to investigate this.

Before setting out with any of these questions the researcher would need to be clear how ‘nursing home’ was to be defined, and for Q3(b) a validated tool to measure knowledge would need to be available. Q3(a) suggests a comparative survey of samples of nursing homes and hospitals, but would the underlying question be answered by asking patients’ views alone? Palliative care is needed up to and after the point of death, and so it might be necessary to extend the survey to satisfaction of next of kin, who can give a full picture of care given. Q3(c) suggests a research design that needs a more in-depth approach, and the answer will be contained in words rather than numbers – Research Example 2.3 (Hasson et al. 2008) describes a study to answer this question using a qualitative approach.

USING A HYPOTHESIS

A hypothesis is a statement that can be tested, and is used mostly in experimental research. Qualitative designs and surveys do not usually have a hypothesis, although sometimes surveys do test for differences between groups and so might use one. Statistics are required to test the hypothesis, which has to be very precisely written. The hypothesis expresses the predicted outcome of the experiment, either in positive or negative terms. As an example, Q2(b) above could be answered by testing a hypothesis, which would be something like the following.

Children under five years of age with reflux given prophylactic antibiotics will experience fewer episodes of urinary tract infection in one year than children with reflux not given prophylactic antibiotics.

The hypothesis might even express the magnitude of the expected difference – in this case, it might be predicted that children given antibiotics will experience, on average, at least 50% fewer infections than those not given antibiotics. But for the purpose of statistical testing, the hypothesis is more often expressed in negative terms, or as a null hypothesis, such as the following example.

Children under five years of age with reflux given prophylactic antibiotics will experience the same number of urinary tract infections in one year as those not given prophylactic antibiotics.

In this case, the experiment would aim to find the null hypothesis false, assuming that prophylactic antibiotics are effective in such cases. Chapter 36 gives more information about how such hypotheses are tested for statistical significance.

SEARCHING AND EVALUATING THE LITERATURE

The next stage is to find out what evidence already exists in the chosen research area. It is a waste of time and money to conduct research where the answer to the question is already known. What is already known about a subject can be found from a variety of sources. Books may be a starting point, but quickly become out of date if the subject matter is topical. Academic journals are a better place to start, and access to online databases such as CINAHL (Cumulative Index of Nursing and Allied Health Literature; see Chapter 6 for more details) make this task speedy and relatively simple. If anything, the problem is that there will be too much information, and Chapter 6 discusses how to refine the search. Beyond written sources, evidence may be found on the internet and various online resources. As well as locating the evidence, it must be appraised and evaluated. Not all that is written is of good quality, and evidence from one country or in one population may not necessarily generalise to other cultures or situations. Chapters 6 and 7 of this book discuss this stage in considerable detail.

Sometimes the research process may consist entirely of a review of the literature. A well-designed systematic review is an accepted research approach in its own right, systematically searching out and evaluating all the research that has been published on a particular topic. In an increasingly complex and fragmented world of information it is important to develop an evidence base that is well validated, and on which practice can be based. Q2(a) above would suggest the need for a systematic literature review, and Chapter 24 deals with this specialised form of research. Research Example 2.4 gives an example of a systematic review.

Most of the questions in the examples above would require a literature review before being able to refine the question further. It might be, for example, that a study has already been conducted testing the effectiveness of pelvic floor exercises in first-time mothers, and found them to be ineffective in reducing time taken in second stage. But can this be applied to women having their second or subsequent child? And can a study conducted in, say, the USA be applied in the UK? A literature search on palliative care in nursing homes might show that nurses in this setting have very low levels of knowledge or interest in palliative care. But the studies are few, out of date and somewhat contradictory. Is it justifiable to conduct a further piece of research in the area?

RESEARCH EXAMPLE 2.4 Systematic Review

Frasure J (2008) Analysis of instruments measuring nurses’ attitudes towards research utilization: a systematic review. Journal of Advanced Nursing 61: 5–8.

This research study used established methods of systematic review to assess instruments that have been developed to measure nurses’ attitudes towards research utilisation. Four electronic databases were searched for relevant articles published during the period 1982 to 2007, and 186 sources were identified. Of these, 25 met the criteria for review, but only 14 were developed with sound psychometric properties. Only one, that by Estabrooks, was found to have been rigorously tested. This instrument was recommended for use, but further work was suggested to develop this area of research.

CHOICE OF METHODOLOGY, RESEARCH DESIGN

The majority of this book (Section 3) is devoted to a description of different research designs. In many ways, the choice of research design is the most important stage of the research process, for it affects all the others. Some questions are more appropriate for an experimental approach; others are entirely suited to an in-depth ethnographic study. Researchers often make explicit a conceptual framework within which they are working, which will determine the overall research approach. A conceptual framework makes clear the researcher’s ‘world view’ – their assumptions and preconceptions about the subject under consideration. In Question 1 above, for example, the researchers may have a conceptual framework that emphasises womens’ right to autonomy in decisions and policies relating to labour. Consequently any research study would be concerned with gathering the experiences and feelings of women about their labour, rather than purely objective clinical data. The kind of data collected, the types of analysis that are possible and the way in which the results can be applied to practice will all depend on the research design.

Some research designs are quantitative. This means they ultimately collect numerical data and are amenable to statistical analysis. Such research designs may or may not have a hypothesis, but experimental studies always require such a statement to be tested statistically. Research Example 2.1 (Salvesen & Morkved 2004) and Research Example 2.2 (Chevalier et al. 2008) both describe quantitative studies. Quantitative designs may be experimental, such as Salveson and Morkved’s design, but may also be observational, such as Chevalier et al.’s survey using a questionnaire. In the latter, structured answers such as ticked boxes enable the data to be coded and translated into numerical form. Surveys may also use medical records or laboratory tests as their data source to estimate the numbers of patients in a community who have measles, for example. Epidemiological studies of the incidence and distribution of diseases also use quantitative methods.

Other research designs are qualitative. These designs use narrative, words, documents or graphical material as their data source, and analyse material to identify themes, relationships, concepts and, in some cases, to develop theory. Such research approaches explore an experience, culture or situation in depth, taking account of context and complexity. Qualitative designs may be used where comparatively little is known about a subject, so no hypothesis can be formulated. The purpose is exploratory rather than explanatory, although qualitative studies may certainly contribute much to our understanding of phenomena and many also develop theory. An example of a qualitative study is given in Research Example 2.3 (Hasson et al. 2008).

Both approaches are valid ways of advancing nursing knowledge. A quantitative study may be very good at finding out the extent of compliance with diabetic therapy, for instance, by measuring levels of the blood glucose in a sample of diabetic patients. A qualitative study, on the other hand, may tell us why it is that certain diabetic patients do not take their insulin as prescribed, by observing and talking to them, and gaining understanding of the context in which the insulin is (or is not) taken.

More than this, qualitative and quantitative methodologies are based on different philosophical assumptions and derive from different historical traditions. Chapter 11 discusses these issues in much more detail, and the reader is encouraged to get to grips with this academic debate. Nursing needs to embrace all research methodologies in order to engage with the breadth of questions that need to be asked. Ours is a discipline drawing on many different traditions of academic enquiry.

The research design (or methodology) is distinct from the methods used for data collection. A single data collection method, for example interview or observation, may be used for many different research designs.

So we can return to our hypothetical questions generated in questions 1 – 3 above, and consider the research methodology that might be appropriate to answer each one. In the example relating to pelvic floor exercises for pregnant women, Q1(a) and Q1(b) are both essentially asking for information that can be gathered in a quantitative survey, but Q1(a) might also be answered by observation of antenatal classes, or examination of the women’s records. Q1(c) will require an experimental design to compare outcomes in two groups (Research Example 2.1). With regard to a potential study examining the prescribing of prophylactic antibiotics, Q2(a) suggests a literature review as described above, but Q2(b) would require a rigorous experimental design to answer the question about effectiveness. Q2(c) requires a survey, as described in Research Example 2.2. Finally, in relation to examining the best way to deliver palliative care in nursing homes, Q3(a) and Q3(b) both suggest a quantitative survey design, but Q3(a) will require a comparative survey, measuring satisfaction in the two types of care setting. It might also be answered using qualitative methods, asking in-depth questions of palliative care patients and their relatives in two types of setting. Indeed, this question might require mixed methods, as discussed in Chapter 27. Q3(c) certainly needs a qualitative approach (Research Example 2.3).

PREPARING A RESEARCH PROPOSAL

Whether a large-scale, multi-centre study costing many thousands of pounds or a small, unfunded study for an educational degree is planned, a formal research proposal is likely to be needed.

Such a proposal is a written statement of what the researcher intends to do, why, how, when and, often, how much it will cost. It is used to gain approval for the research, secure funding if it is required, and then to guide the research process during its execution. It will often be modified in the light of pilot studies or practical difficulties, but it is important that the detailed intentions are clear at the outset. It has been said that if you don’t know where you are going you are unlikely to get there!

Chapter 8 sets out the content of a research proposal in detail, but the precise form of the proposal will vary according to the nature of the research and the purpose of the written proposal. A proposal written in response to a funding call from the National Institute of Health Research or the Medical Research Council is likely to be a substantial document of many pages, written by a team of experienced researchers. One written for the purpose of outlining a small study for a master’s degree may be only a few pages, written by the postgraduate student themself with some guidance from their supervisor.

Whatever the context, however, the proposal will certainly include a section on each of the stages of the research process outlined in Table 2.1. It will also include a section detailing the ethical issues raised by the research, and how the researcher will ensure that confidentiality, informed consent and other ethical principles are respected. Chapter 3 discusses these issues in more detail. It is usual to include a table or Gantt chart showing the timescale of the project. Table 2.2 shows such a chart for a complex evaluation study involving a survey, documentary analysis, case studies and focus groups. It is also helpful to identify milestones, stating the date by which each stage of the research will be completed, though this is obviously subject to change as the inevitable obstacles and delays come into play. It is customary to include a breakdown of resources required and a justification of why they are needed.

Table 2.2 Example of a Gantt chart for a mixed method piece of research

Clearly, the research proposal cannot be written until the researcher has thought through all the stages of the research process in some detail. However, the proposal is of necessity one of the early stages in the process, as it is impossible to proceed without one.

GAINING ACCESS TO THE DATA

Because of the sensitivity of much of the research that takes place in healthcare, and the vulnerability of many of its subjects, a complex system of governance has been developed in the UK to ensure all research is approved for its ethical soundness, scientific quality and legal propriety. NHS trusts are also concerned to ensure that all research that takes place within the trust is properly funded and insured against liability. A system of ethical regulation via the National Research Ethics Service (www.nres.npsa.nhs.uk/) is in place, and all applicants carrying out research in healthcare must follow this system. In addition, since 2001 a system of research governance has been developed to guard against research that has not been properly scrutinised and approved, after various high-profile scandals concerning NHS research (Department of Health 2005).

Chapter 10 deals with this topic in depth. Suffice to say at this stage that the system is necessary, but rather bureaucratic and time-consuming. Depending on arrangements at each local trust, it is likely to take anything from 4 to 20 weeks from completing a research proposal to having all the required permissions in place to begin data collection (Gerrish & Guillaume 2006).

In addition to formal permission, however, access to the data may require negotiation of a more informal nature with local personnel who act as ‘gatekeepers’. If access to patients or their records is needed, for example, it may be necessary to gain the co-operation of the appropriate consultant, practice manager or audit department in addition to ethical and research governance committees. Access to a nursing home or school will require the permission of the appropriate senior manager. Chapter 10 also deals in more depth with this informal process of negotiating access.

SAMPLING

Once the research begins, the first stage is likely to be selecting the sample. Unless it is a complete census, researchers collect data from a selected group, rather than an entire population. In our earlier examples, samples might be taken from antenatal class attenders, nursing homes in a particular region of the country, consultants in paediatric medicine or relatives of patients requiring palliative care. How are the samples to be selected, and how many is enough? These questions are dealt with in detail in Chapter 12, but the answers are rarely simple, particularly about sample size.

A quantitative study involving a comparison between two groups is likely to